Data cleaning method for demand prediction of shared automobile

Abstract

The invention relates to a data cleaning method for demand prediction of a shared automobile. The data cleaning method comprises the following steps: (1) extracting opening data of all client applications interacted at the same time; (2) acquiring all network points within a 5-kilometer range of an interaction place and data about whether to-be-rented vehicles exist at the corresponding network points or not when the interaction occurs at this time; (3) obtaining data about whether an order placing operation is carried out or not after the interaction; (4) allowing the process id for opening the client application at this time to belong to a network point closest to the interaction place; (5) obtaining the longest vehicle taking walking distance of successful order placing of the user whoperforms the interaction at this time; (6) obtaining information of all network points within the longest vehicle taking walking distance of the user; (7) judging whether a to-be-rented vehicle existsor not; and (8) affiliating the data result of the interaction to the network point closest to the interaction place, and finishing data cleaning. The data validity is ensured; dirty data cannot be doped, prediction accuracy is improved, and meanwhile model operation time is shortened.

Description

technical field [0001] The invention belongs to the field of shared cars, and in particular relates to a data cleaning method for demand prediction of shared cars. Background technique [0002] "Car sharing" first appeared in the 1940s and was invented by the Swiss. They have organized "self-driving car cooperatives" across the country, which is very practical in mountainous countries like Switzerland. After one person finishes using the car, handing over the car keys to the next person is easier than building a network in flat countries. [0003] Later, Japan, the United Kingdom and other countries rushed to follow suit, but none of them took shape. Japan mainly because the automakers don't support the plan, the Japanese like to have a personal car of their own. In the UK, despite government support, the cost of car rental is low, which hinders the development of "car sharing". [0004] At present, many companies have shared cars on the streets of China, but the current ...

AI Technical Summary

Problems solved by technology

[0005] 1. The validity of the data cannot be guaranteed;

[0006] 2. It will be mixed with dirty data, which will affect the accuracy of prediction

Images